Background: Glycans are synthesized in the Golgi by sequentially acting glycosyltransferases, but it is not known how their functions are coordinated in live cells. Results: N-and O-glycosyltransferases form enzymatically active homo-and/or heteromeric complexes. Conclusion: Glycosyltransferases function as physically distinct enzyme complexes rather than single enzymes. Significance: The results help understand the overall functioning of the Golgi glycosylation pathways both in health and disease.

Smoking decreases the synthesis rates of type I and III collagens in skin in vivo and alters the balance of extracellular matrix turnover in skin.

Acidic pH of the Golgi lumen is known to be crucial for correct glycosylation, transport and sorting of proteins and lipids during their transit through the organelle. To better understand why Golgi acidity is important for these processes, we have examined here the most pH sensitive events in N-glycosylation by sequentially raising Golgi luminal pH with chloroquine (CQ), a weak base. We show that only a 0.2 pH unit increase (20 microM CQ) is sufficient to markedly impair terminal alpha(2,3)-sialylation of an N-glycosylated reporter protein (CEA), and to induce selective mislocalization of the corresponding alpha(2,3)-sialyltransferase (ST3) into the endosomal compartments. Much higher pH increase was required to impair alpha(2,6)-sialylation, or the proximal glycosylation steps such as beta(1,4)-galactosylation or acquisition of Endo H resistance, and the steady-state localization of the key enzymes responsible for these modifications (ST6, GalT I, MANII). The overall Golgi morphology also remained unaltered, except when Golgi pH was raised close to neutral. By using transmembrane domain chimeras between the ST6 and ST3, we also show that the luminal domain of the ST6 is mainly responsible for its less pH sensitive localization in the Golgi. Collectively, these results emphasize that moderate Golgi pH alterations such as those detected in cancer cells can impair N-glycosylation by inducing selective mislocalization of only certain Golgi glycosyltransferases.

Altered glycosylation has turned out to be a universal feature of cancer cells, and in many cases, to correlate with altered expression or localization of relevant glycosyltransferases. However, no such correlation exists between observed enzymatic changes and the expression of the oncofetal Thomsen-Friedenreich (T)-antigen, a core 1 (Gal-beta1 --> 3-GalNAc-ser/thr) carbohydrate structure. Here we report that T-antigen expression, instead, correlates with elevated Golgi pH in cancer cells. Firstly, using a Golgi-targeted green fluorescent protein (GT-EGFP) as a probe, we show that the medial/trans-Golgi pH (pHG) in a high proportion of breast (MCF-7) and colorectal (HT-29, SW-48) cancer cells is significantly more alkaline (pHG > or = 6.75) than that of control cells (pHG 5.9-6.5). The pH gradient between the cytoplasm and the Golgi lumen is also markedly reduced in MCF-7 cells, suggesting a Golgi acidification defect. Secondly, we show that T-antigen expression is highly sensitive to changes in Golgi pH, as only a 0.2 pH unit increase was sufficient to increase T-antigen expression in control cells. Thirdly, we found that T-antigen expressing MCF-7 cells have 0.3 pH units more alkaline Golgi pH than non-expressing MCF-7 cells. Fourthly, in all cell types examined, we observed significant correlation between the number of T-antigen expressing cells and cells with a markedly elevated Golgi pH (pHG > or = 6.75). Consistent with these observations in cultured cells, cells in solid tumors also heterogenously expressed the T-antigen. Thus, elevated Golgi pH appears to be directly linked to T-antigen expression in cancer cells, but it may also act as a more general factor for altered glycosylation in cancer by affecting the distribution of Golgi-localized glycosyltransferases.

Bullous pemphigoid (BP) is a subepidermal blistering skin disease, which has shown a strong association with neurological diseases in epidemiological studies. The BP autoantigens BP180 and BP230 are expressed in the cutaneous basement membrane and the central nervous system. Using BP180 and BP230 ELISA assays and immunoblotting against BP180, we analyzed the IgG reactivity in the sera of 115 patients with Alzheimer's disease (AD) and 40 neurologically healthy controls. BP180 autoantibodies were found in 18% of patients with AD, whereas only 3% of controls had positive results (P = 0.019). BP230 values were higher and more often elevated in patients with AD than controls, but not significantly. None of the positive AD sera that recognized the full-length human BP180 in immunoblotting reacted with the cutaneous basement membrane in indirect immunofluorescence analysis. Moreover, a retrospective evaluation of the hospital records of the patients with AD revealed neither BP diagnosis nor BP-like symptoms. Interestingly, increased BP180-NC16A autoantibody values correlated with cognitive decline measured by mini-mental state examination scores, but not with the concentration of AD biomarkers in cerebrospinal fluid. Our findings further the understanding of the role of BP180 as a shared autoantigen in neurodermatological interactions and the association between BP and neurodegenerative diseases.